---
id: "claim:verifying-llm-extracted-text-with-token-alignment-2026-02-10-biorxiv:4b9e517c3078ec4906bc"
kind: "claim"
title: "BOAT downstream analyses restrict targets to multi-word phrases."
description: ""
url: "https://sina.bio/claim/boat-downstream-analyses-restrict-targets-to-multi-word-phrases"
---

# BOAT downstream analyses restrict targets to multi-word phrases.

## Content

BOAT downstream analyses restrict targets to multi-word phrases.

## Metadata

```json
{
  "claim_position": 140,
  "source_publication_id": "version:verifying-llm-extracted-text-with-token-alignment--2026-02-10-biorxiv",
  "source_work_id": "work:verifying-llm-extracted-text-with-token-alignment"
}
```

## Connections (1)

### Verifying LLM-extracted text with token alignment

- ID: `version:verifying-llm-extracted-text-with-token-alignment--2026-02-10-biorxiv`
- Kind: `publication`
- Direction: `incoming`
- URL: <https://sina.bio/publication/verifying-llm-extracted-text-with-token-alignment>

This publication contains this grounded claim.

> In downstream analyses we restricted to multi-word targets to focus on phrases where ablating an interior token introduces a meaningful non-contiguity.

Source: 2026-02-10-biorxiv.pdf — <https://sina.bio/source/2026-02-10-biorxiv-pdf>
